Maintenance and estates
Memory does not scale past one kitchen. How to move an estate from Saturday-night emergencies to one boring calendar - registers, schedules and the 70:30 ratio.
Maintenance and estates
The ops manager of a ten-site group can usually tell you, to the pound, what last month's emergency call-outs cost. What they often cannot tell you is when each site's extraction was last cleaned, which walk-ins are on their original compressors, or why site four gets through door seals like consumables. That asymmetry is the tell: the estate is being run reactively, and reactive is the expensive setting - emergency work typically costs three to five times the planned equivalent, before counting the lost trade while the kit is down.
Multi-site maintenance fails differently from single-site. One kitchen with a diligent head chef can run on memory and habit. Ten kitchens cannot, because memory does not scale, habits vary by site, and the person who knows the quirks of unit seven's combi hands in their notice eventually. An estate needs the knowledge moved out of heads and into a system.
The framework
Every plannable thing an estate does hangs off knowing what it owns: each appliance and plant item per site, with age, condition and service history. The register is what turns "site four's fridge is playing up again" into "unit F4-03 is nine years into a 10-15 year life and on its third seal - budget replacement next quarter". It is also where statutory and quasi-statutory items - gas safety, electrical testing, extraction cleaning, LEV examination - stop being remembered and start being scheduled.
Facilities management solved this decades ago. SFG20, published by BESA since 1990, is the recognised UK standard library of maintenance specifications - more than 1,500 schedules setting out what each asset type needs and how often, updated hundreds of times a year as legislation moves. An estate does not need to adopt the whole apparatus; it needs its plan to borrow the principle that every asset class has a defined routine at a defined frequency, done by a defined competence. The recognised good-practice budget shape that follows is roughly 70:30 planned to reactive - if your invoices run the other way, the plan exists on paper only.
The estates that make this work treat compliance items as the spine of the calendar, because they move the least and cost the most to miss. Extraction cleaning is the canonical example: TR19 Grease intervals follow each site's cooking hours, every completed clean generates the certificate the insurer expects, and the schedule logic is the same whether you run two kitchens or forty - we set it out in how often to clean a kitchen extraction system. Group the sites onto one supplier and one renewal cycle and the admin collapses to a single diary line per service, which is how hotels run deep cleaning across multiple outlets without the calendar owning them.
Making it stick
The whole estate's planned work - servicing, statutory tests, extraction cleans, deep cleans, filter changes - lives in one shared calendar with one owner, booked a year ahead in each site's quiet windows. Preventive programmes are consistently found to cut maintenance costs by around 12-18% against run-to-failure, but the operational win is bigger than the invoice line: planned work happens at 7am on the quiet Tuesday, reactive work happens at 7pm on the busy Saturday. Seasonal estates get an extra gift here - golf clubs and leisure caterers can stack the heavy work into the closed season and enter peak trade with everything certificated.
Reactive jobs never reach zero, and they should not be wasted when they happen. Every call-out is data: log it against the asset, and patterns surface fast. The estate that reviews its call-out log quarterly finds the recurring offenders - the site whose water chemistry eats dishwashers, the walk-in that needs its plant replaced rather than re-gassed again - and converts them into planned lines before they convert themselves into Saturday emergencies.
Acquisitions test the system, and the plan should have a standard answer ready. A new site joins the estate with unknown kit and usually incomplete paperwork; the onboarding routine is a condition survey into the asset register, the statutory spine booked immediately - gas, electrics, an extraction clean with certificate if none is in date - and its servicing merged onto the group calendar and group suppliers. Done in the first month, it costs a survey and a few certificates; left a year, the site quietly imports its previous owner's reactive habits into your invoice mix.
Questions
The recognised good-practice budget shape in facilities management is roughly 70:30 planned to reactive. No estate reaches zero reactive work - equipment fails, accidents happen - but an estate spending most of its maintenance money on emergencies is paying the 3-5x premium that unplanned work carries, plus the lost trade. The ratio is worth tracking quarterly from invoices: it is the single clearest health metric a maintenance plan has.
Yes - they should be its spine. Gas safety, electrical testing, extraction cleaning to TR19 Grease and LEV examination carry fixed or evidence-based intervals, generate the certificates insurers and landlords ask for, and cost the most when missed. Scheduling them first, per site, and hanging routine servicing around them gives the calendar a structure that survives staff turnover - which is the failure mode that kills most multi-site plans.
Phoenix Duct Clean · by the numbers
TR19 Grease extraction cleaning across your whole estate - intervals banded per site, scheduled in your quiet windows, certificates delivered to one inbox.